Reporting guidelines: CONSORT, STROBE, PRISMA, STARD
- Concept
Reporting guidelines are lists of what a paper must report for its design, so that the reader can judge it.
What it is
Reporting guidelines are checklists, often with a flow diagram, that set out what a paper should report for its design. CONSORT is for randomised trials, STROBE for observational studies (cohort, case-control, cross-sectional), PRISMA for systematic reviews and STARD for diagnostic accuracy studies. They improve the transparency of reporting; they are not tools for rating study quality.
Worked example
- Data (made up): the CONSORT diagram of a trial; 640 people assessed for eligibility.
- Excluded: 140 (did not meet criteria or declined); randomised: 640 − 140 = 500.
- Allocation: 250 to the intervention group and 250 to control.
- Lost to follow-up: 20 in intervention (8 %) and 30 in control (12 %).
- Intention-to-treat analysis: 250 and 250; per protocol: 250 − 20 = 230 and 250 − 30 = 220.
- Reading: the diagram shows at a glance how many were lost in each arm and who was analysed.
Which guideline for which design
| Guideline | Design | Flow diagram |
|---|---|---|
| CONSORT | Randomised controlled trial | Assessed, randomised, allocated, lost, analysed |
| STROBE | Cohort, case-control, cross-sectional | Recommended to show participants at each stage |
| PRISMA | Systematic review and meta-analysis | Records identified, screened, excluded and included |
| STARD | Diagnostic accuracy | Participants by index test and reference standard result |
| CARE | Case report | Not applicable (case timeline) |
| SPIRIT | Trial protocol | Schedule of enrolment, interventions and assessments |
Markers
- CONSORT
- STROBE
- PRISMA
- STARD
In practice
- When reading a trial, the CONSORT diagram is the fastest place to check losses and intention-to-treat analysis.
- Many journals ask for the matching checklist when a manuscript is submitted.
Common pitfalls
- Using CONSORT or STROBE as a quality score: they are reporting guidelines, not risk-of-bias tools.
- Choosing the wrong guideline: STROBE for a trial or CONSORT for a cohort.
Test yourself
Think of the answer, then tap the question to see it.
Which reporting guideline fits a case-control study?
STROBE, which covers observational studies: cohort, case-control and cross-sectional.
Which design is PRISMA for?
Systematic reviews and meta-analyses.
Does meeting CONSORT guarantee that a trial is of good quality?
No: it guarantees that it is well reported; quality is judged with risk-of-bias tools.
Connections
Educational content on epidemiological and statistical methods; the examples use made-up numbers to teach the calculation. Not a clinical or legal guide: current Colombian rules are in the official sources.
Sources: CONSORT 2025 Statement: updated guideline for reporting randomised trials (EQUATOR Network, 2025); The PRISMA 2020 statement: An updated guideline for reporting systematic reviews (EQUATOR Network, 2021).
Spotted a mistake?We are always improving for the community. Tell us what is wrong and our team will review it.Report a mistake
Representative model, for educational purposes only.